Analyzing Public Sentiment Towards LLM: A Twitter-Based Sentiment Analysis
Niharika Prasanna Kumar, Kishore Srinivasan, Dhanesh Ramesh · 2023
In this mixed-method research of tweets from December 2022 to April 2023 mentioning LLM, to optimize the data collection and analytics methods to focus on critical aspects. The set of information includes tweet content, usernames, locations, and sentiment indicators (positive tweets or negative tweets). The exploration into the Large Language Model’s (LLM) early users places an emphasis on the location details thereby providing insights about user spread without resorting to follower counts or user descriptions. On top of that, the study incorporates the establishment of major topics with the help of topic modeling techniques and analysis of sentiments linked with these tweets. Sentiment labels (positive or negative) are included as a distinct column within the dataset. Subjective insights into early users’ viewpoints and understandings are assembled through manual sampling and thematic segregation of tweets. Eventually, this refined approach allows for a comprehensive assessment into LLM’s reception, proficiency, and hurdles during the stated time period all while staying true to ethical practices regarding data gathering